Summary

Continued discussion from previous week addressing:

• Recap of decisions we made last week re: one-time import of Tasks and how Chandler will map dates
• Date/time entry widget proposals. See DateTimeProposal
• Sharing of personal alarm settings, which lead to a revisiting of our favorite topic:
• What are the primary use cases for calendar sharing?
• The difference between sharing and import/export, from the user's perspective.
• Recurrence: how complex will it be for Kibble? If it's simplified, how do we deal with more complex recurrence from other clients?
• Status update on free/busy

Attendees

• JeffreyHarris
• AlecFlett
• SheilaMooney
• MimiYin

Notes

• Recap of decisions we made last week re: one-time import of Tasks and how Chandler will map dates
• No regular syncing of Tasks with non-Chandler clients
• On one-time import of Tasks into Chandler, users are given a choice of:
  • Selecting one date attribute (Start date, Due date or Alarm) to a Chandler alarm date OR
  • Putting tasks with Start dates and/or Due dates on the calendar as (multi-day) all-day task/events
  • And mapping Alarm dates to Chandler alarm dates
• Mimi: I'm wondering if we can just have some smart defaults under the hood. Basically decide for the user. If there's just an Alarm, we match it to a Chandler alarm. If there's a Start date and/or Due date, we put it on the calendar.

• Date/time entry widget proposals. See DateTimeProposal
• Concern was raised over whether users would understand whether they could skip the Start time field and head straight to the End date field in order to make multi-day, all-day events.

• Sharing of personal alarm settings
• Not allowed in shared collections
• However, users will be allowed to export alarm settings

• What are the primary use cases for calendar sharing?
• Mitch and Esther
• Blue and OSAF
• Design team calendar
• But perhaps sharing your own calendar with yourself on another Chandler client (ie. on your home computer) is the most wide-spread, common denominator calendar sharing use case?

• The difference between sharing and import/export, from the user's perspective.
• It seems like in order to satisfy the "share your own calendar with yourself at home on Chandler" scenario, users would have to use the export/import feature (in order to share alarms), which feels harder to do than simply sharing the collection with themselves (which would not share alarms).
• What do we do about this seeming incongruity between how common a scenario is (sharing calendars with yourself) and how easy or obvious it is to do it (export/import).
• [OI^?] Can we share alarms with a sharee if the sharee is also the sharer?
• [OI^?] Can we model that as simply an ACL level that is automatically set for the user under the hood? No UI necessary.

• Recurrence: how complex will it be for Kibble? If it's simplified, how do we deal with more complex recurrence from other clients?
• We had proposed only allowing for the simplest kind of recurrence (to limit UI work to a simple pulldown)
  • Daily
  • Weekly
  • Monthly on this date
  • Monthly on this day of the week
  • Yearly
• If users receive events from other clients via sharing that have more complex recurrence formulas, Chandler will throw up an error dialog
• Jeffrey commented that: Weekly every MWF was quite common for a lot of people, so we may want torevisit this issue for .7

• Status update on free/busy
• Jeffrey to read Lisa's proposal
• We're not going to have a free/busy attribute on the calendar that is orthogonal to event status for Kibble.

Next steps

• Mimi to come up with UI proposal for one-time import of tasks